Kentucky’s graduated driver licensing program for
young drivers: barriers to e
Vective local
implementation
L C Steenbergen, P S Kidd, S Pollack, C McCoy, J G Pigman, K R Agent
Abstract
Objectives—To examine the
implementa-tion of graduated driver licensing (GDL) in Kentucky and to use the data collected to recommend actions to enhance the eVectiveness of the GDL program.
Methods—Data were acquired from
sur-veys of 700 law enforcement oYcers and more than 40 judges and from interviews with 100 persons who implement or are aVected by Kentucky’s GDL program—for example, traYc court judges, licensing clerks, law enforcement oYcers, insur-ance agents, driving instructors, parents, and employers of teens. Transcripts from interviews were analyzed using a qualita-tive data analysis computer program.
Results and conclusions—Participants
noted a widespread lack of awareness of the night-time driving restriction and a substantial number of young drivers re-ceiving little driving time during the learner permit phase. It appeared that specific GDL provisions can be diYcult for judges and law oYcers to enforce and the penalty of license suspension after several traYc violations may not be a suf-ficient deterrent. EVorts are needed to increase parental awareness of GDL pro-visions, GDL purpose, and their teen’s traYc violations and to increase parental enforcement of restrictions that are diY-cult for law enforcement agencies to monitor, such as the night-time driving restriction and the adult supervision re-quirement.
(Injury Prevention 2001;7:286–291)
Keywords: graduated driver licensing; teen driver; motor vehicle crashes; adolescent mortality
The involvement of teenaged drivers (16–19 years old) in motor vehicle crashes continues to be a serious problem in Kentucky as well as in most of the developed world. In 1999, Kentucky’s teenage crash death rate was almost three times the rate for other licensed drivers. During 1995–99, teen drivers were associated with 24% of injury crashes in the state, despite representing only 6.4% of licensed drivers.1 Thirty two states have attempted to address this problem though graduated driver licensing (GDL) programs.2
This legislation extends the supervised learning period to improve driving skills and decision making, imposes restrictions to help protect young drivers from hazardous situations while they learn to drive, and applies penalties to motivate young drivers to drive
safely. Requiring new drivers to demonstrate safe driving (by having no traYc violations) before progressing to a less restricted level is a key element of most GDL programs.3
This requirement is not, however, a part of Ken-tucky’s program. Examination is needed of the policy, its implementors, and the settings in which decisions and actions occur to identify impediments to successful implementation.4 Lessons learned in this process evaluation may be applicable to other policy measures in other jurisdictions.
Kentucky’s partial graduated licensing program
Kentucky’s GDL program includes: (a) a six month learning permit level (may start at age 16) that requires adult supervision; (b) a restriction on driving after midnight during the permit level; (c) a six point limit on traYc vio-lations to age 18, with a penalty of license sus-pension; and (d) a four hour driving education class. In addition, blood alcohol concentration (BAC) limit is 0.02 ml/dl for drivers under age 21. There are no special limits on the number or age of passengers. Kentucky’s program no longer meets the minimum requirements for a “full GDL” program under current National Highway TraYc Safety Administration guide-lines. Missing are three provisions: (a) a visibly distinguishable intermediate level license, (b) a limit on unsupervised night-time driving in the intermediate stage, and (c) a requirement to be free of traYc violations for a period of time before graduating to the next level of licensure.5
EVect on teen driver crashes
Researchers at the Kentucky Transportation Center and Kentucky Injury Prevention and Research Center have examined teen crash data for the three years (1993–95) before and three years (1997–99) after Kentucky enacted its program (1996).1
Results indicate a 32% reduction in crash rates for 16 year old drivers due to an 83% decrease in crash rates for driv-ers age 16 to 161⁄2, the age that drivers are in
the adult supervised permit level. Unfortu-nately, there have been no decreases in the number of crashes for drivers over age 161⁄2.
Therefore, although Kentucky’s program has saved lives, reduced injuries, and reduced crash costs for permit age drivers, there is no current evidence that Kentucky’s “partial GDL” pro-gram has suYciently addressed the teen crash problem. Results from this study indicate a need for additional measures to decrease motor Kentucky Injury
Prevention and Research Center, 333 Waller Ave, Suite 202, Lexington, KY 40504–2915, USA L C Steenbergen College of Nursing, Arizona State University, Phoenix, AZ P S Kidd Departments of Pediatrics and Preventive Medicine, Kentucky Injury Prevention and Research Center, University of Kentucky, Lexington, KY S Pollack School of Nursing, Northern Kentucky University, Highland Heights, KY C McCoy Kentucky Transportation Center, College of Engineering, University of Kentucky, Lexington, KY J G Pigman K R Agent Correspondence to: Lorena Conroy Steenbergen [email protected]
vehicle crashes in the 161⁄2to 18 year old age
group.1
Research evaluating GDL programs in most jurisdictions indicates that teen crashes have decreased significantly after implementation of GDL programs.5–14
However, studies also indi-cate that the eVect on 17 and 18 year old driv-ers varies greatly among jurisdictions, from a substantial reduction (17% in Nova Scotia6
), to an increase (6% in Kentucky1
). The reason for this variation has not been determined.
The study
The purpose of this study was to examine the perspective of participants who are responsible for the implementation of Kentucky’s GDL or are aVected by the program, and to use the data collected to recommend actions to enhance the eVectiveness of this and similar programs.
This report describes methods and results of a survey of 700 law enforcement oYcers and more than 40 district judges conducted in 2001 and of a qualitative analysis of transcripts from interviews with 100 participants conducted in 1999. The interviews were conducted in four Kentucky counties representing urban and rural areas from four geographical regions.
Methods
QUALITATIVE DATA COLLECTION:INTERVIEWS
Sampling procedures
Groups were identified who were responsible for local implementation of the GDL program or who would be expected to be aVected by it (primary participants). These included driver educators, district judges, police chiefs, sher-iVs, licensing clerks, emergency department supervisors, insurance agents, and employers from randomly chosen businesses. Parents and teen drivers (secondary participants) were ran-domly selected from those working in the
agencies, businesses, hospitals, or schools where interviews with primary participants had been conducted. Structured interviews by one person were conducted with the 100 partici-pants shown in table 1 and 87 were audiotaped. Detailed field notes were taken on sessions with those who refused. Teens were given question-naires to be mailed back anonymously regard-ing knowledge, compliance, and attitude to-ward GDL to supplement the qualitative interviews.
Qualitative data analysis
Transcripts from the interviews were entered into the QRS NU*DIST ([Non-numeric Unstructured Data Index Searching and Theo-rizing] Rev 4; Qualitative Solutions & Research Pty Ltd, Melbourne, Victoria, Australia) com-puter program for coding and sorting. Text was coded line by line and analyzed to identify recurring themes. Coding for statements from the 20 teens interviewed was compared with data from completed questionnaires from these same teens and no discrepancies were noted. Hard copies of five transcripts were used to assess inter-rater reliability that showed 99% agreement in category coding patterns.
QUANTITATIVE DATA COLLECTION:SURVEYS
Sampling procedures
Questionnaires were developed based on issues identified in the interviews and from analysis of crash data. These were sent to all state law enforcement posts and 400 were returned. In addition, 300 city and county level law enforcement oYcers attending randomly se-lected mandatory in-service training classes also completed questionnaires. A separate questionnaire was sent to judges in 59 districts; 43 judges from 34 districts completed the questionnaires.
Results INTERVIEWS
Themes from interviews were generated based on groupings of related categories with “sub-stantial coding”. The four primary themes relating to implementation are listed in table 2.
Enforcement diYculties
Statements from all subgroups, except employ-ers and health care providers, identified problems with enforcement relating to alcohol, the six point violation limit, the night-time Table 1 Interview participants
Interview participants
County
Total
A B C D
Judges and judicial support staV 2 4 3 2 11 Court clerk/licensing 2 2 2 2 8 Law enforcement oYcers 4 2 6 2 14 Driver instructors/educators 4 2 1 1 8 Insurance agents 3 2 3 2 10 Emergency health care 2 1 0 1 4 Employers of teens 3 3 2 3 11 Parents of teen GDL drivers 4 3 3 4 14 Teen GDL drivers 5 5 5 5 20
Total 29 24 25 22 100
Table 2 Frequency of themes in interviews
Major theme Lines oftext Subgroups most often verbalizing theme
Enforcement diYculties 500 54% of judicial 71% of law enforcement 75% of educators 70% of teens
Problems with penalties 900 21% of parents 100% of insurance agents 73% of judges 86% of law enforcement
50% of driver educators Lack of knowledge (to varying degrees) regarding GDL/provisions 900 73% of judicial 93% of law enforcement 86% of parents 88% of driver educators 75% of teens 80% of insurance agents Recommendations for education 1300 73% of judicial 86% of law enforcement 79% of parents 100% of driver educators 65% of teens
driving restriction, and the adult supervision requirement.
Inadequate quantity and quality of driving instruction—Teens and driver education
in-structors indicated that most teens they knew were not getting adequate driving experience or quality driving instruction from their parents or guardians during the six month per-mit phase. Suggestions included requiring the parent/guardian to certify that the teen has received a specified number of hours of driving practice during the instructional permit level before issuing license.
Primary enforcement—OYcers reported they
believed they did not have primary enforce-ment rights—that is, were not able to legally stop a driver who is suspected of being in viola-tion of the night-time driving restricviola-tion or adult supervision requirement for permit driv-ers. They stated they believed this did not con-stitute “probable cause” (that is, a legal justifi-cation for making a traYc stop) because permit status could not be determined until after the oYcer stopped the driver and checked the operator’s license. If the driver proved they had a regular license, rather than a learner’s permit, the initial cause for the stop would become invalid because the restriction would not apply. OYcers expressed the concern that, in these cases, any subsequent charges made after the traYc stop—for example, open alcohol con-tainer, alcohol on breath, drug paraphernalia— might be dismissed in court. Some oYcers suggested that a “decal” or placard would help identify a vehicle driven by a permit driver (as is customary in some other countries) and thereby facilitate enforcement of permit level driving restrictions.
Uninformed parents—Additional problems
identified through judicial comments included lack of parental awareness of the teen’s citations, and the absence of parents in the courtroom for these oVenses. They added that a parent may not know of violations or “points” accumulated by their teen until they received notice of license suspension.
Parents not enforcing GDL provisions—The
majority of judicial and teen participants, and all law enforcement participants, stated that they knew of parents who were not enforcing GDL provisions, especially the night-time driving restriction. They recommended steps to increase parents’ awareness, motivation, accountability, and responsibility for enforcing GDL provisions, especially the night-time driving restriction.
Range of enforcement—Judicial and law
en-forcement participants expressed concern that teen licenses in Kentucky could be easily altered to allow underage purchase of alcohol. Responses indicated there was perception of a wide range of enforcement of the alcohol (0.02 ml/dl BAC) provision, from strict to lenient. Teens, law enforcement oYcers, and judicial participants reported few to zero citations were given for violations of the night-time driving restriction provision (or adult supervision requirement), except in the county that already had an established teen curfew law.
Fines and suspensions
Participants also identified problems and made recommendations related to imposing fines and license suspension penalties for violations.
Problem with license suspension—Responses
from the judicial branch indicated that the penalty of suspended license that can be administratively imposed for exceeding the six point traYc violation limit often resulted in the additional problem of older teens driving with-out a license and losing their insurance. Persons were later seen in court on subsequent violations for driving without an operator’s permit or driving without insurance, thus pro-ducing a greater burden on the court system. Because the first traYc violation is usually deferred, teens may have four moving viola-tions without having their license suspended.
EVect of penalties on others—Comments
indi-cated the greatest impact of current Kentucky GDL penalties—that is, fines and suspensions—was often on the parent, not on the teen. The majority of insurance agents indicated that a license suspension that came to their attention would significantly influence (the parent’s) automobile insurance coverage and often resulted in minimal liability cover-age. Insurance agents stated that if a teen driver was involved in a vehicle crash while driving on a suspended license, an insurance claim might be denied.
Lack of knowledge
Seventy four participants, representing all four counties and all subgroups, made statements indicating lack of awareness regarding one or more of the GDL provisions, GDL purpose, or indicating misconceptions regarding the law, especially the night-time driving restriction. Comments from teens, parents, and licensing clerks indicated that the parent/guardian is often not aware of the contents of the GDL orientation and driving instruction booklet which the teen and guardian are given when applying for a driving permit. Many were also not aware that they were expected to provide extensive driving instruction to their teen. In a direct observational study of the permit process in all four counties, no licensing clerk in any county was seen instructing parents regarding GDL provisions or to read the booklet given to the teen driver.
Recommendations for education
Statements from the majority of participants identified a need for increased education. These views were expressed in response to the question “What else can Kentucky do to help decrease the number of teen crashes in Kentucky?” Table 3 summarizes the recom-mendations made; these are listed in more detail in the survey results section. Law enforcement oYcers and the judicial subgroup indicated that increased public and parental awareness might help facilitate judicial and law enforcement eVorts, which may in turn, increase compliance.
JUDICIAL AND LAW ENFORCEMENT SURVEYS
Survey results revealed substantial support for the primary issues identified in the interviews. More than 93% of the 700 law enforcement respondents felt that Kentucky should make legislative changes to address repeat teen oVenders of Kentucky’s (0.02 BAC) alcohol law, make teen driver’s licenses less easily altered, and require parents to be notified of a teen’s traYc violations. More than 90% of respondents noted substantial unawareness of the night-time driving restriction and more than 80% noted a problem with the penalty of license suspension. In addition, results indi-cated that three fourths of respondents felt that is it diYcult to enforce a night-time driving restriction for permit drivers. Half did not think there was legal justification for stopping a driver who is suspected of violating the night-time driving restriction for permit level drivers. More than 80% felt specific edu-cational eVorts would help improve the pro-gram. Law enforcement survey results are summarized in tables 4 and 5.
Discussion
This research supports the current literature indicating that enforcement, compliance, awareness, and education all influence the eVectiveness of injury prevention programs. However, several issues identified in this proc-ess evaluation are unique. Findings relating to penalties specific to this evaluation include the problem of license suspension as the primary penalty, and the problem of penalties not hav-ing suYcient eVect on the teen driver. These problems may not be as significant with other GDL programs because many jurisdictions do not use license suspension as the primary pen-alty. Instead, they require a clean driving record (free of crashes/convictions) for a period of time before progressing to the next GDL level or full licensure.
ENFORCEMENT AND COMPLIANCE
Although the diYculty for law enforcement agencies to enforce the night-time driving restriction and the adult supervision require-ment is a unique finding, previous studies have Table 3 Educational recommendations: summary
Recommendation Lines of text Primary concern
Education for law enforcement and judicial subgroup regarding GDL provisions and enforcement
100 lines Law enforcement Education for parent and communities regarding teen crash problem,
GDL provisions, and parental role in enforcement
331 lines Law enforcement Parents Increased teen education regarding driving skills, decreasing risks 1017 lines All groups
Table 4 Results from law enforcement oYcer questionnaires: per cent agree (average)
Comments % Agree
1. Many people do not seem know about the GDL night-time driving restriction for permit drivers 93 2. The penalty of license suspension on teen drivers can result in persons being seen back in court for
driving without a license and driving with no insurance
83 3. The teen’s parent/guardian is usually not required to accompany the teen in court for a traYc violation 82 4. In many cases, the teen’s parent/guardian may not know of the teen driver’s traYc violations 91 5. It is hard to enforce the GDL night-time driving restriction for permit drivers 73 6. A suspicion of violation of the night-time driving restriction does not necessarily constitute “probable
cause” for stopping a vehicle
62
Note: average per cent from state level law enforcement and city/county level law enforcement surveys. Table 5 Recommendations from law oYcer questionnaires: per cent support (average)
Recommendation % Support*
1. Include a GDL provision to allow stronger penalties for teen drivers repeatedly convicted of DUI violations (0.02 BAC law)
96 2. Allow a teen’s record of previous DUI violations to be accessible to the courts (BAC 0.02 to 0.08
ml/dl)—allows identification of repeat oVenders 95 3. Make driver licenses for those under age 21 more diYcult to alter/fabricate to help decrease purchase of
alcohol by minors using altered identification
93 4. Require licensing agency to notify parents/guardians of teen’s traYc violations 93 5. Provide increased teen EDUCATION on driving safety 91 6. Provide EDUCATION to the law enforcement oYcers, judges, and prosecutors implementing GDL. This
includes providing clear information on GDL purpose and provisions, their specific role in enforcement of the provisions, ideas on eVective ways to enforce the provisions, and statistical results of their eVorts.
86
7. Allow courts to require the presence of parents (guardians) when teens are in court for traYc violations 91 8. Regularly disseminate local and statewide teen crash statistics 83 9. Conduct a MEDIA campaign to encourage parents to enforce provisions such as the “night-time driving
restriction” and “adult supervisor/escort requirement” for permit drivers
82 10. Conduct a MEDIA campaign focusing on the first six months of independent teen driving, to inform
parents about the need for close monitoring, skills training, and need for limit on passengers and driving privileges
82
11. Require vehicles driven by new teen drivers to display a sign or decal identifying them as a novice or permit drivers under GDL provisions and restrictions (336 asked)
53
*Note: “support” is defined as responses of “definitely would help”, “would help”, “might/should help” improve Kentucky’s GDL program. Average of percentages between the state level law enforcement and city/county level law enforcement. DUI = driving under the influence.
identified diYculties in enforcement of provisions,15–18
non-compliance with the night-time driving and passenger restrictions,1 15
and inadequate driving experience for some permit level drivers. Compliance varies with the particular restriction, with the licensing level, and the jurisdiction.15 19–21
Analysis of Kentucky crash data involving drivers age 16 to 161⁄2after
GDL verifies that there is substantial non-compliance with permit level provisions, such as the night-time driving restriction and the adult supervision requirement.1
Research in New Zealand indicates that non-compliance and low level of oYcial enforcement have a sig-nificant influence on GDL eVectiveness.15 22 Survey results in Ontario and Nova Scotia revealed that a substantial number of permit drivers (15% to 30%) reportedly drove only two or less times per month in the learning stage of GDL.19 20
Twenty eight states have addressed this issue by requiring a minimum number of hours of supervised driving during the learner stage.2
LACK OF KNOWLEDGE AND EDUCATIONAL NEEDS
The specific recommendations for GDL re-lated education for law enforcement oYcers and the judges are unique findings, although lack of knowledge of GDL and recommenda-tions for parent, teen, and community educa-tion for injury preveneduca-tion programs have been previously documented.4 23–28
A lack of aware-ness of the penalties for violations of the GDL restrictions was especially noted among the teens surveyed in Nova Scotia.19
In Kentucky, survey results indicated that only 50% of the participants were aware of GDL and related restrictions for teen drivers.29
Participants recommended providing in-creased teen education aimed at improving teen driving safety/skills and motivating teens to drive more safely either through GDL in school programs or though driver education courses. The literature does not clearly support the participants’ confidence in driver education courses in the high schools as a means to help improve teen driving or decreasing risks.30 Other studies have identified the importance of providing local injury data to communities to increase the motivation for compliance with injury prevention policies and to increase sup-port from families, populations, and authori-ties.23 24
Dowswell et al reported that legislation without a strong educational component might result in poor compliance or ineVective imple-mentation,31
and Cassady et al have described the importance of increasing awareness through media to the success of injury preven-tion programs.4
LIMITATIONS
Three judges in one county and five law enforcement oYcers in another county re-quired the interviews to be conducted as a group, despite having scheduled individual interviews for all adults. Group bias may have occurred in these groups in which senior or supervisory members were present. Lines of text coded from the interviewer’s notes attrib-uted to the participants who refused to be
audiotaped may under-represent the discourse that actually occurred. Responses do not necessarily represent non-working parents, non-working teens, or employers in small, independently owned businesses.
Implications for prevention
There is widespread lack of knowledge regard-ing the night-time drivregard-ing restriction and limited parental enforcement of some GDL permit provisions, such as adult supervised driving instruction. This may be resulting in insuYcient driving experience and inadequate protection from risks for some teens during the permit stage. EVectiveness of Kentucky’s program may also be impaired due to di Ycul-ties in enforcement of some permit provisions. Non-cumulative penalties for teen alcohol (0.02 BAC) cases, and license suspension for exceeding the six point traYc violation limit (plus deferred violations) may not have su Y-cient deterrent eVect. Research in Canada has revealed that parents feel the night-time driving restriction is easy to monitor and enforce.19 20 Therefore, eVorts should be made to increase parental enforcement of restrictions that are diYcult for law enforcement agencies to moni-tor, such as the night-time driving restriction and the adult supervision requirement. In addition, measures are needed to increase the hours of driving practice for permit level driv-ers.
Specific legislative enhancements and edu-cational components are supported by the findings in this study. Kentucky, and other jurisdictions implementing GDL programs, should consider the specific recommendations listed in table 5 and address issues identified in table 4 by participants in this study to increase the eVectiveness of the program.
In addition, the results indicate that jurisdic-tions with GDL programs should conduct process evaluations at intervals to help identify and remove barriers to eVective local imple-mentation of the GDL program. Qualitative research is an important component of policy process evaluations and can help identify prob-lems with enforcement, awareness, and compli-ance with an existing policy as well as help identify legislative changes and educational eVorts needed to help maximize the benefits of the injury prevention program. Follow up sur-veys can help to validate and prioritize issues identified though interviews.
Issues eVecting local implementation of the GDL program
x Parental compliance with providing suY-cient driving experience for new drivers. x Parental enforcement of GDL
restric-tions.
x Awareness of GDL purpose, provisions and role.
x Enforceability of GDL provisions. x EVectiveness of GDL penalties.
x Education for all those implementing the policy.
1 Agent K, Steenbergen LC, Pigman J, et al. Impact of the partial graduated drivers license program in Kentucky on teen motor vehicle crashes. Transportation Research Record 2001 (in press).
2 Insurance Institute for Highway Safety. US licensing systems for
young drivers. Highway Loss Data Institute (http//: highwaysafety.org/safety_facts/state_laws/grad_license.htm). 3 University of North Carolina, Highway Safety Research
Center. Graduated driver licensing overview, CB# 3430 (http://www.hsrc.unc.edu/pubinfo/grad_overview.htm). 4 Cassady CE, Orth DA, Guyer B, et al. Measuring the
imple-mentation of injury prevention programs in state health agencies. Inj Prev 1997;3:94–7.
5 National Highway TraYc Safety Administration, US Department of Transportation. Saving teenage lives: the case
for graduated driver licensing. Washington, DC, 1998.
6 Mayhew DR, Simpson HM, des Groseilliers M. Impact of
the graduated driver licensing program in Nova Scotia.
Ottawa, Ontario: TraYc Injury Research Foundation, 1999.
7 Ulmer RG, Preusser DF, Williams AF, et al. EVect of
Florida’s graduated licensing program on the crashes of teenage drivers. Arlington, VA: Insurance Institute for Highway
Safety, 1999.
8 Jones B. The eVectiveness of provisional licensing in Oregon: an analysis of traYc safety benefits. J Safety Res 1994;25:33–46.
9 Ulmer RG, Preusser DF, Ferguson SA, et al. Teenage crash reduction associated with delayed licensure in Louisiana. J
Safety Res 1999;30:31–8.
10 Boase P, Tasca L. Graduated licensing system evaluation:
interim report ’98. Toronto, Ontario: Safety Policy Branch,
Ministry of Transportation of Ontario, 1998.
11 Langley JD, Wagenaar AC, Bragg DJ. An evaluation of the New Zealand graduated drivers licensing system. Accid
Anal Prev 1996;21:33–46.
12 National Highway TraYc Safety Administration, US Department of Transportation. State legislative fact sheets:
graduated driver licensing system. 1999 (http://www.nhtsa.
dot.gov/people/outreach/stateleg/graduated.html). 13 Foss RD. Preliminary evaluation of the North Carolina
gradu-ated driver licensing system: eVects on young driver crashes.
University of North Carolina, Highway Safety Research Center, 2000 (http://www.hsrc.unc.edu/pdf/2000/gdl/ pdf.pdf).
14 Ulmer RG, Ferguson SA, Williams AF, et al. Teenage crash
reduction associated with delayed licensure in Connecticut.
Insurance Institute for Highway Safety (http//www. highwaysafety.org).
15 Begg DJ, Langley JD, Reeder AI, et al. The New Zealand graduated driver licensing system: teenagers’ attitudes towards and experiences with this car driver licensing system. Inj Prev 1995;1:176–81.
16 Preusser DF, Ulmer RG, Preusser CW. Obstacles to
enforcement of youthful (under 21) impaired driving.
Washing-ton, DC: National Highway TraYc Safety Administration, US Department of Transportation, 1992.
17 Whines JR. Young people and driving: the graduated driver
licensing system. National road safety seminar. Wellington,
New Zealand. Seminar papers 1988: 325–45.
18 National Highway TraYc Safety Administration, US Department of Transportation. Review of the literature
evaluating the eVect of countermeasures to reduce alcohol impaired driving (1980–1989). Final report. Volume II— Individual analyses and assessments. (DOT HS 808 024.)
Washington, DC: US Department of Transportation, 1991.
19 Mayhew DR, Simpson HM, Ferguson SA, et al. Graduated licensing in Nova Scotia: a survey of teenagers and parents.
J TraYc Med 1998;26(1–2):37–44.
20 Mayhew DR, Simpson HM, Ferguson SA, et al. Graduated licensing in Ontario: a survey of parents. J TraYc Med
1999;27(3–4):71–80.
21 Frith WJ, Perkins WA. The New Zealand graduated driver licensing system. National road safety seminar. Wellington, New Zealand. Seminar papers 2, 1992: 256–78. 22 Williams AF. Graduated licensing comes to the United
States. Inj Prev 1999;5:133–5.
23 Kemp A, Gibbs N, Vafidis G, et al. Safe Child Penarth: experience with a Safe Community strategy for preventing injuries to children. Inj Prev 1998;4:63–8.
24 Coggan C, Disley B, Patterson P. Community based intervention on adolescent risk taking: using research for community action. Inj Prev 1998;4:58–62.
25 Agent KR, Pigman JG. Analysis of traYc accident data in Kentucky 1994–1998. (KTC-99–56.) Kentucky
Transpor-tation Center, University of Kentucky, 1999.
26 Evans L. TraYc safety and the driver. New York: Van Nostrand Reinhold, 1991
27 COMSIS Corporation and the Johns Hopkins University.
Understanding youthful risk taking and driving: interim report.
(Contract No DTNG22–93-C-15182.) September 1995, Maryland (www.nhtsa.dot.gov/people/injury/research). 28 Klitzner M, Vegega M, Gruenewald P. An empirical
examination of the assumptions underlying youth drinking/ driving prevention programs. Evaluation and Program
Plan-ning 1988;11:219–35.
29 Agent KR, Pigman JG. Executive summary: traYc safety issues opinion survey. Kentucky Transportation Center,
Univer-sity of Kentucky, 1999.
30 Vernick JS, Guohua K, Ogaitis S, et al. EVects of high school driver education on motor vehicle crashes, violations, and licensure. Am J Prev Med 1999;16:40–6.
31 Dowswell T, Towner EML, Simpson G, et al. Preventing childhood unintentional injuries—what works? A literature review. Inj Prev 1996;2:140–9.